November 8th 2016
By David Speights, Ph.D., Daniel Downs, Ph.D., and Adi Raz, MBA
Merchandise return transactions are a critical part of the customer experience of retail. The objective is an optimal return rate, where the retailer finds the right balance between too many returns that may lead to fraud and abuse and too few returns that may lead to customer unhappiness. But how does one arrive at such a measure?
A major retailer recently asked the same question. As a customer of Appriss Retail, the retailer asked for help in understanding the impact return policies have on business. The retailer, who was experiencing declining sales, needed to pinpoint areas where they might improve, and one of the areas to be evaluated was the return counter.
With return authorization solutions deployed in more than 27,000 retail locations, Appriss Retail evaluates an unprecedented amount of return data, and in turn, is able to provide its customers with powerful insights to help them understand and improve their business. Therefore, Appriss Retail set forth on a benchmark study to examine the impact different return policies could have on net sales and return rate.
Stores were divided into three groups:
The “strict return rules” test group and the “friendlier return experience” Appriss Retail test group were the most easily compared since they rolled out their new return procedures at the same time. And the results speak for themselves.
Over the course of the test, which ran for six months, the Strict Group showed an 11.2 percent decrease in net sales, while the BOC stores showed a 6.4 percent decrease in net sales. And the Friendly Group showed only a 2.6 percent decrease in net sales—this is an 8.6 percent improvement over the Strict Group.
Taking a closer look, the customer evaluated the net sales trend by store group, comparing the time in the months leading up to the deployment date with the months following. During that time, the new Friendly stores showed a minimal change. The net sales decrease leading up to the deployment date was 0.1 percent, while after the deploy date it was 0.2 percent. However, the net sales decrease in the stores where a “strict” policy was implemented had a 0.6 percent net sales decrease leading up to the change, and a whopping 2.1 percent decrease in the months after they adopted the strict return rules policy.
Appriss Retail dug deeper to determine whether these trends differed based on region, and the decline in the Strict stores was universal and widespread across all tested regions.
Appriss Retail also evaluated the impact the stricter return policies and Friendly stores had on return rate. Based on the findings, the Friendly stores and Strict stores both had better return rate reductions than the Balance of Chain control group. The Friendly stores exhibit return rate reduction due to Appriss Retail’s targeted predictive modeling approach that only impacts a small portion of all returns (typically less than 2 percent). The Strict stores exhibit return rate reduction using more punitive rules and policies, for example, by declining all non-receipted returns, and directly impacting the overall consumer experience of every shopper.
At the end of the six-month test period, the Strict stores showed an 8.6 percent reduction in net sales compared to Friendly sales. Stated a different way, if sales declined in the entire chain by 8.6 percent and the overall return rate was 7.2 percent, the retailer would have to reduce their total return dollars by approximately 118 percent to offset the loss in revenue – which is impossible; the total dollars of lost sales would have already exceeded the total return dollars for the retailer.
This shows a real revenue impact and telling statistics for the retailer and a key learning point for other retailers. Additional important conclusions drawn from the study include:
David Speights, Ph.D., is the chief data scientist, Daniel Downs, Ph.D., is a statistical criminologist, and Adi Raz, MBA, is director of modeling and analytics for Appriss Retail.